This paper extends paradigm of stroke monitor aimed to make visble acute ischemics signs in unenhanced CT brain imaging. Method based on soft tissue extraction followed by multiscale processing of tresholded areas for hypodensity extraction and adaptive visualization of underlined symptoms was optimized with segmentation of stroke susceptible regions. Regular, compact regions were segmented in three-stage algorithm based on noise reduction, adaptive seeds selection, and region growing with adaptive mean and standard deviation criteria. Experimental verification confirmed important role of improved segmentation for increased efficiecny of stroke monitor.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.